A technical case study comparing Nike and New Balance highlights a growing AEO risk tied to performance, not rendering theory. Nike’s site uses Next.js with server-side rendering, but depends on a massive ~9MB JavaScript hydration bundle to become interactive. For Googlebot, this is likely mitigated through dynamic rendering and whitelisting. For newer AI agents and generic LLM scrapers, it is not.
Those non-whitelisted bots receive the full payload, often timing out or failing to process the page efficiently. New Balance, by contrast, serves fully populated server-side HTML with immediate content availability and no hydration gap. As search shifts from a single crawler to thousands of AI agents, architectures that rely on whitelisting become fragile. The takeaway: fast, plain HTML is the safest default for future AI visibility.
